Skip to main content

Neurovn SDK and CLI for tracing agentic AI workflows

Project description

Neurovn SDK

Python SDK and CLI for tracing agentic AI workflows.

Installation

The package is published on PyPI as neurovn.

python -m venv .venv
source .venv/bin/activate
pip install --upgrade neurovn

Verified release: neurovn==0.1.1.

If you are contributing from this monorepo and need an editable install:

cd neurovn-sdk
pip install -e .

Point the SDK at the hosted Neurovn backend:

export NEUROVN_API_URL=https://agentic-flow.onrender.com

Quick Start

CLI Integration

NEUROVN_API_URL=https://agentic-flow.onrender.com neurovn trace ./workflow.json --workflow-name "My Workflow" --canvas-name "My Workflow" --source cli

Module entrypoint fallback:

NEUROVN_API_URL=https://agentic-flow.onrender.com python -m neurovn trace ./workflow.json --workflow-name "My Workflow" --canvas-name "My Workflow" --source cli

Decorator Integration

from neurovn import trace

@trace.agent(name="Research Agent", model="gpt-4o", provider="OpenAI")
async def research(query: str):
    response = await openai.chat.completions.create(...)
    return response

@trace.tool(name="Web Search", tool_id="mcp_web_search", tool_category="mcp_server")
async def web_search(query: str) -> str:
    return "search results"

with trace.session("My Workflow", source="decorator", canvas_name="My Workflow"):
    result = await research("latest AI updates")

if trace.last_result:
    print(trace.last_result["canvas_id"])

Features

  • Trace Decorators: Instrument Python functions with @trace.agent and @trace.tool
  • CLI: Emit workflow JSON files to Neurovn backend
  • Session Management: Group multiple calls into a workflow session and print a terminal summary for explicit sessions
  • Usage Tracking: Capture token usage from LLM responses

Documentation

For full integration guides and examples, visit Neurovn Docs.

Notes

  • External users should install from PyPI. The editable install path is contributor-only.
  • neurovn trace ... and python -m neurovn trace ... are equivalent entrypoints.
  • Local trace persistence depends on the backend being reachable and configured with a valid Supabase service-role key.

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

neurovn-0.1.1.tar.gz (13.0 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

neurovn-0.1.1-py3-none-any.whl (10.0 kB view details)

Uploaded Python 3

File details

Details for the file neurovn-0.1.1.tar.gz.

File metadata

  • Download URL: neurovn-0.1.1.tar.gz
  • Upload date:
  • Size: 13.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.10 {"installer":{"name":"uv","version":"0.10.10","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for neurovn-0.1.1.tar.gz
Algorithm Hash digest
SHA256 dfd390fcabbc4a79900a76c86f3ad6703e4e05bb150984a28b79db72b4a440ca
MD5 4007614c831cbd526530cb7e6c63ff80
BLAKE2b-256 27bb11027bb116f6e545b99c4c282095b93b5f751ea61980654eebce6374b247

See more details on using hashes here.

File details

Details for the file neurovn-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: neurovn-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 10.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.10 {"installer":{"name":"uv","version":"0.10.10","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for neurovn-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 d16dffd49d0567dbcbf8dfb7a844583de6b9a09292537f69f455b533b0ba39ea
MD5 270d135f6e559684703fa3bac2c303b6
BLAKE2b-256 c32f6edfe4156940f10f7a8587b7d2edf078c37827cfe6219160a343d164d4ca

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page